The mortality from HIV-1 infection has been greatly reduced by the introduction of antiretroviral treatment (ART). ART, however, often fails due to the existence of drug resistant viruses. Drug resistance occurs because of the presence of viral variants with mutations in the genes encoding the targeted enzymes. The frequency with which these drug resistant viruses preexist prior to treatment is unclear. Within an individual patient, the generation of these drug resistance mutations presumably depends upon the initial virus population, mutation frequency, viral fitness, and selection pressure. Much of our understanding of drug resistance comes from investigations of virus populations under strong selection pressure of ART. These previous studies have clearly demonstrated that continued viral replication in the face of drug therapy often leads to drug resistant viruses. It remains unclear whether these viruses typically exist prior to ART and whether there are factors that would help predict their existence. In order to understand these questions we propose to analyze samples from an ART naive cohort of patients in Africa. We hypothesize that drug resistant viruses will exist prior to drug therapy. In addition, we predict that virus populations with greater genotypic diversity in the pol/pro locus, which encodes the enzymes, reverse transcriptase (RT) and protease, targeted by ART, will have a greater likelihood of having drug resistant viral variants. Understanding the factors influencing the frequency and persistence of drug resistant viral mutants prior to drug therapy will be important for improving the clinical care of patients with HIV/AIDS. Furthermore, ART is beginning to be widely introduced in Africa often without rigorous clinical monitoring or advice. With these studies, we will gain insight into the likelihood of success with ART in Africa and the possibility of the evolution of a 'new' epidemic with drug resistant HIV -1 viruses.